Prevalence, incidence and predictors of severe anaemia with zidovudine-containing regimens in African adults with HIV infection within the DART trial.

Objective: To describe the prevalence, incidence and predictors of severe anaemia in previously untreated symptomatic HIV-infected adults with CD4+ T-cells <200 cells/mm(3) initiating zidovudine-containing regimens in Africa.

Design: DART is a randomized trial comparing two strategies for HIV/AIDS management in Uganda and Zimbabwe.

Methods: We analysed the occurrence of anaemia at weeks 4 and 12, and then every 12 weeks. We also evaluated sex, age, WHO stage, body mass index (BMI), baseline laboratory measurements and first regimen as predictors of developing grade 4 anaemia (<6.5 mg/dl) by week 48 using logistic regression.

Results: To May 2005, 3,314 participants (65% women, 23% at WHO stage 4, median age=37 years, baseline CD4+ T-cell=86 cells/mm(3) and median baseline haemoglobin=11.4 g/dl) had a median 72 weeks follow-up. Prevalence of grade 4 anaemia was 0.70, 2.0%, 0.5% and <0.5% at weeks 4, 12, 24 and > or =36, respectively. Overall, 219 (6.6%) participants developed grade 4 anaemia by week 48; women and those with lower haemoglobin, CD4+ T-cell count and BMI at baseline were at significantly higher risk (P<0.05), but not those with lower neutrophils or receiving cotrimoxazole at baseline.

Conclusions: We observed a higher incidence of grade 4 anaemia than in studies from industrialized countries, which is likely to be due in part to population characteristics and in part to a higher rate of concurrent HIV-related clinical events. Clinical vigilance and haemoglobin measurements 4, 8 and 12 weeks after starting zidovudine could help to manage serious anaemia.